Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 18 Apr 2022 14:36:56 -0700
From:   Eric Biggers <ebiggers@kernel.org>
To:     Nathan Huckleberry <nhuck@google.com>
Cc:     linux-crypto@vger.kernel.org,
        Herbert Xu <herbert@gondor.apana.org.au>,
        "David S. Miller" <davem@davemloft.net>,
        linux-arm-kernel@lists.infradead.org,
        Paul Crowley <paulcrowley@google.com>,
        Sami Tolvanen <samitolvanen@google.com>,
        Ard Biesheuvel <ardb@kernel.org>
Subject: Re: [PATCH v4 6/8] crypto: x86/polyval: Add PCLMULQDQ accelerated
 implementation of POLYVAL
Message-ID: <Yl3Z+CqZIZOBTfOs@sol.localdomain>
References: <20220412172816.917723-1-nhuck@google.com>
 <20220412172816.917723-7-nhuck@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220412172816.917723-7-nhuck@google.com>
Precedence: bulk

A few more comments:

On Tue, Apr 12, 2022 at 05:28:14PM +0000, Nathan Huckleberry wrote:
> +/*
> + * Computes the 256-bit polynomial represented by LO, HI, MI. Stores
> + * the result in PL, PH.
> + *   [PH :: PL] = [HI_1 : HI_0 + MI_1 :: LO_1 + MI_0 : LO_0]
> + */

It is unclear what the double colon means.  Maybe you meant for it to indicate
128-bit boundaries, as opposed to 64-bit boundaries?  It is not used
consistently, though.

> +/*
> + * Computes the 128-bit reduction of PL : PH. Stores the result in dest.

This should use the order "PH : PL", to be consistent with the notation
elsewhere.

> diff --git a/arch/x86/crypto/polyval-clmulni_glue.c b/arch/x86/crypto/polyval-clmulni_glue.c
[...]
> +struct polyval_ctx {
> +	/*
> +	 * These powers must be in the order h^8, ..., h^1.
> +	 */
> +	u8 key_powers[NUM_PRECOMPUTE_POWERS][POLYVAL_BLOCK_SIZE];
> +};
> +
> +struct polyval_desc_ctx {
> +	u8 buffer[POLYVAL_BLOCK_SIZE];
> +	u32 bytes;
> +};

As I've mentioned elsewhere, it is confusing to have both ctx and desc_ctx.  The
former should be called polyval_tfm_ctx, like it is in polyval-generic.c.

> +asmlinkage void clmul_polyval_update(const u8 *in, struct polyval_ctx *keys,
> +				     size_t nblocks, u8 *accumulator);

The argument order here is a bit weird.  It would be more logical to have it be
(keys, in, nblocks, accumulator), similar to crypto_shash_digest().

Also, 'keys' should be const.

- Eric